Jasper Hong Sim ^[1], Department of English Language and Literature, National Institute of Education, Nanyang Technological University of Singapore

Abstract

Singapore English (SgE) is one of the post-colonial varieties of English that has been regarded as having reached endornormative stabilisation. However, it is clear that perspectives of what is standard and what is colloquial can be different; in the domain of language teaching, this affects how SgE is used, taught and treated. Because SgE is inherently a new English, the questions of which particular norms have indigenised and stabilised, the degree to which SgE has stabilised in its entirety, and what different communities of people regard as norms, can be complex. Though many studies have been carried out on how the colloquial variety is being treated by language teachers in the classroom, such complexity has been neglected.

This article aims to provide an insight into this issue through a feature-based approach. Twenty English trainee teachers were surveyed, and responses were collected to 13 items of SgE features in terms of grammaticality/acceptability, appropriateness, and likelihood of corrections. The results show that features were indeed treated differently, and they fall into three main categories. It also revealed that some features which are deemed ungrammatical in Standard English were being treated as formal and grammatical.

Keywords: Singapore English, Singlish in the classroom, English-language teaching, acceptability of Singlish, English trainee teachers.

Introduction

In the description of post-colonial varieties of English by Schneider (2003: 263), the Dynamic Model regards Singapore English (SgE) as an established variety that has reached the stage of 'endonormative stabilisation'. This is generally defined by the simultaneous adoption and acceptance of SgE as an indigenous linguistic norm; a 'symbolic expression of the pride of Singaporeans in their nation', as well as the distancing of the colonial British English (BrE) (Schneider, 2007: 106). However, because the model is defined by stages of continua, and not by discrete stages of succession, the term 'stabilisation' is therefore relative. That is, because SgE is inherently a new English, the questions of what particular norms have indeed indigenised and stabilised, the degree to which SgE has stabilised in its entirety, and what different communities of people regard as norms, can be complex.

A pertinent issue is how this complication manifests itself in the language classroom. Particularly in the debate between the standard variety, Singapore Standard English (SSE), and the basilectal variety, Singapore Colloquial English (SCE), teachers are inundated with educational initiatives that are intertwined with intricate tensions of, on one hand, the teaching of an English that is internationally acceptable (Goh, 1999), and on the other, managing a variety that 'qualifies as a dialect facilitating emotional expressiveness and play – an identity carrier' (Schneider, 2003: 265). The shadow of the colonial variety, British English (henceforth referred to as Standard English,^[2] StdE), too, plays a part in this complex issue. Clearly, teachers' perspectives of SSE, SCE, and StdE can be different, and this affects how languages are used, taught and treated. Singapore does not have a governing body such as the Académie française for the French language or the Dewan Bahasa dan Pustaka for the use of the Malay language in Malaysia and Brunei; rather, 'general goals and guidelines of language-related policies are expressed in policy speeches by political leaders' (Kuo and Jernudd, 1994: 31). The Ministry of Education is most directly involved in matters relating to language policies, planning, and implementation processes. If language education is significantly influential in the linguistic landscape and plays a pivotal role in the acquisition and development of English (Goh and Silver, 2006), then the nature of what constitutes 'English' in the language classroom by different teachers needs substantial definition. In this regard, past research that has attempted to delineate this issue has mainly explored SgE as an entirety (e.g. Tay, 2000; Farrell and Tan, 2007), but clearly, this cannot give a full account because of the heterogeneity that exists within SgE.

Literature review

Background to the SCE-SSE debate

Since the inception of nationhood in 1965, linguistic issues have always been a political discourse in Singapore. Tong and Pakir (1996: 182) explained that because of the geographical, socio-cultural and political nature of Singapore, social policies were founded upon the cornerstone of an ideology of pragmatism, rather than on emotionally laden values of religion and culture (Lee, 2011). The government needed to prepare a populace that caters to the demands of the world economy, for its survival relies very much on its usefulness to major powers (Gopinathan, 2007: 58). Educational initiatives are therefore found imbued with such intricate tension (Rubdy and Tupas, 2009: 318). In the same vein, English has become the 'workhorse of economic capital' (Alsagoff, 2010: 341) for its value as the language of international diplomacy, finance, science and technology. The populace is concomitantly impelled toward the acquisition and use of an English that is internationally acceptable, as foregrounded in many of the political speeches (Goh, 1999; Lee, 1999; Lee, 2011) and articulated in the English Language Syllabuses (MOE, 2001: 3; MOE, 2011: 7).

In this regard, the existence of SCE becomes problematic. This is because whereas SSE conforms generally to Standard English^[3] in terms of vocabulary and grammar (Low and Brown, 2005: 11-12), SCE differs in many aspects as a result of substrate-influenced nativisation^[4](Gupta, 1994; Platt and Weber, 1980). Therefore, the government sees SCE as pernicious and malignant, and endeavours to eradicate it (Chng, 2003; Wee, 2005). This was manifested most in the launch of the Speak Good English Movement (SGEM), which was a campaign through which the use of SCE was 'rectified' to one that is 'grammatically correct [and] universally understood' (SGEM, n.d.). However, the resistance efforts by the government such as SGEM have caused considerable debate in schools and in the media.^[5]

In the context of language education

Studies on English language teaching within the SSE-SCE debate can be generally subsumed as apropos to these two broad categories – the cognitive aspects and/or the practice of teachers, which are what Borg (2003: 81) encompassed in his notion of 'teacher cognition'. Research on cognitive aspects encompasses that of beliefs, (e.g. teachers' perception of SCE/SSE/StdE, appropriateness judgments), and knowledge (e.g. theories, competency in distinguishing between StdE and SSE/SCE features and grammaticality judgements of SCE features). Studies on practices include, for instance, frequency of corrective feedback, or corrective strategies employed. Certainly, most studies explored more than one aspect. In our discussion, however, it is expedient to discuss previous studies concerning these aspects separately, although they are not independent entities.

Numerous studies have attempted to ascertain teachers' perception and beliefs of SCE in the language classroom. Saravanan and Poedjosoedarmo (1997) surveyed 326 trainee teachers, and they concurred that, generally, only SSE is appropriate for the classroom but SCE is permissible exceptionally for constructing interpersonal relationships. Teh (2000) also surveyed the beliefs of 20 teachers and found similar findings, although the teachers did see SCE as possible impediment to SSE acquisition and writing performance (p. 49).

Research germane to aspects of knowledge has been scarce. Schaetzel et al. (2010) ascertained the performance of English trainee teachers in distinguishing SSE-SCE features before and after an academic module on Singapore English, and concluded that more extensive training has to be done in this respect. However, the subjects were able to differentiate approximately two thirds of the items containing SCE features for the pre- and post-tests. Also, English teachers have varied views on what is grammatical in SSE vis-à-vis StdE. In early scholarship of Singapore English, Tay (1993) asserted that the standard variety that is most practical is one that is endornomative, internationally intelligible, yet highly reflective of local characteristics. However, there is still no absolute definition of what constitutes SSE, as one would have for British/American English, because such stabilised norms have not been coded or developed (Tan, 2011: 6). Therefore, for some, 'a staff is sick', 'furnitures', or 'fish and chips are…' may be acceptable in SSE (Low and Brown, 2005: 59), and some regard them as features of SCE, whereas for others, they may remain unduly prescriptive to rely on an exornomative standard (i.e. native varieties). A teacher participant in Farell and Tan (2008: 387), for instance, had asserted that American English was not StdE, and connotatively regarded 'color' as linguistically substandard.

Finally, several studies on teachers' practices were also done in recent years, but are considerably inadequate. Studies include both correction and usage. Findings on corrections describe two aspects of practice: (1) what is corrected, and (2) how frequently. Farrell and Tan (2007), for instance, found that three teacher participants tended not to correct all instances of SCE, and feedback was given at varied frequencies, and on different aspects of SCE (p. 391). Regarding usage, Gwee's (2011) corpus-based study of 20 primary-school teachers found that SCE was used almost twice more frequently by content-subject teachers for classroom access and management than in English language lessons (p.122). The survey results of Tan and Tan (2008: 473) also showed that students are more likely to use SCE with Mathematics teachers and during Mathematics classes.

Significance of ascertaining aspects of teacher cognition

The findings from these studies have important implications for English language education in Singapore. For instance, research has shown that SCE potentially influences SSE (e.g. on the acquisition of grammar (Dixon, et al., 2012)), which could inform initial teacher education on, for example, strengthening knowledge of systematic differences between SCE and SSE to instruct learners explicitly. Conversely, contrary to the hegemonic argument by scholars and politicians that SCE only impedes the acquisition of SSE (Chng, 2003: 56; Goh, 1999; Lee, 2001), many other studies have posited the pedagogical value of SCE. These studies propounded that SCE can help establish and construct teacher-student interpersonal relationships (Rubdy, 2007: 322; Farrell and Tan, 2007: 389; Schaetzel et al. 2010: 428; Teh, 2000: 30; Kwek, 2005: 81), and could aid in the acquisition of SSE (Rubdy, 2007: 320; Teh, 2000: 30). Similarly, this can inform initial teacher education so that teachers are more aware of the social and functional value of SCE as a classroom resource; and to appreciate the language classroom as for the acquisition of a second language, and not for acts of some form of deficiency rectification (Schaetzel, et al., 2010: 429).

Differential treatment of SgE

From these studies it is clear that:

1. teachers hold varied beliefs and knowledge of what is regarded as grammatical;
2. some are still relying on an exonormative standard;
3. correction of SCE occurs at different frequencies;
4. not all ungrammatical features are corrected;
5. not all features of SCE are allowed in the classroom.

From these findings we can conclude that English language teachers treat SgE, especially SCE, differently. In this regard, though much has been done to investigate the teaching of English within the SCE-SSE debate, most studies observe the treatment of SgE often as a monolithic entity, and this may not adequately account for such differences. That is, few feature-based studies have been done in this area that ascertain what particular features of SgE are regarded as (un)grammatical, (un)acceptable, or will (not) be corrected. Further, most studies measured one or two of the three aspects of teacher cognition without considering the intricate, tripartite relationship between beliefs, knowledge, and practices as part of 'teacher cognition'.

This study, then, aims to evaluate teachers' treatment of SgE with regard to these three aspects by adapting a feature-based approach. It aims to find out:

1. What do teachers regard as grammatical, and to what extent are they grammatical in their own everyday speech?
2. To what degree would a particular feature be appropriate for the classroom?
3. What is the likelihood of this item being corrected by the teacher if it were to be used by students during lesson time?

Methodology

Participants

The scale of this study and the make-up of the English-language teacher population posed considerable constraints on the selection of participants. Although practising teachers may be more reflective concerning what occurs in the language classroom, the complicated and extensive administrative procedures required by the ministry and schools made engaging them unfeasible for this three-month-long study. A second consideration is that most practising teachers are graduates of the Post Graduate Diploma in Education (PGDE) programme in NIE who hold degrees from diverse disciplines, with most not trained in linguistics, or even in the technical aspects of English (a full list in MOE, 2013). In view of this, the varied academic backgrounds and constraints rendered the procuration of practising teachers, or trainees from the PGDE, impractical for the scale of the study. The other group of possible participants were trainee teachers from the Bachelor of Arts/Science (Education) programme in NIE who have English Language as any one of their two Academic Subjects^[6] (AS). Within their academic structure, these English majors/minors are trained in the linguistic aspects of English, including a foundational module in StdE grammar entitled 'AAE104 Exploring Grammar of English', and obliged to take a sociolinguistic module, entitled 'AAE201 The Social Variation of Language'. They also undertake pedagogical modules. With regard to in-school practice, these trainees have had seven weeks of practicum at local schools which requires them to teach. These participants are, therefore, both familiar with the practical issues of English education and the latest educational initiatives, as well as concurrently still strongly connected with linguistic perspectives and knowledge.

Therefore, the participants selected were 20 third-year trainee teachers from the Bachelor of Arts/Science (education) programme who had English Language as one of their two ASs. Given the limitations of this study, these participants were selected based on convenience sampling, and therefore restricted to a small sample size. Consequently, to preserve the internal validity of this study, participant characteristics, such as language background, learning experience, and language proficiency are matched to the fullest possible extent. All participants were Singaporeans who had undergone the same English curriculum prior to their entry into the university. They were also assumed to possess strong command of SSE, as NIE students are required to have at least a 'B3' at the GCE 'O' Level examination, 'B' for General Paper at the GCE 'A' Level examination, or have at least passed the English Language Entrance Proficiency Test prior to admission.

Methodological approach and instrument

Each participant was asked to complete an online questionnaire comprising 26 English sentences (see Appendix A); half of them contained well-established features of SgE grammar and vocabulary, while the other half contained distractors that conformed to norms found in StdE. An example of each is shown below.

My other brother in National Service. (SgE)

Weather permitting, we shall have an open-air concert. (StdE)

For each item, participants were required to indicate their responses to the grammaticality, appropriateness in the classroom, and likelihood of feedback. Grammaticality refers to whether a construction adheres to the rule of English, appropriateness refers to attitudes on whether a feature is more/less apt for the classroom, while likelihood of feedback refers to whether the participant would correct the usage of the particular feature in the classroom. These were clearly defined and described prior to the survey (see Appendix B).

As the notion of grammaticality and acceptability is complex, the measure for grammaticality is such that acceptability was taken into account. That is, if the item was deemed ungrammatical, subjects have to decide on the item's acceptability for use in formal/informal contexts (e.g. ungrammatical but acceptable in informal contexts). In this regard, Gasser (2003) proposed that grammaticality refer to conformity with language rules, while acceptability, then, refers to whether one is pragmatically acceptable in a given context. Therefore in this study, grammaticality was measured in binary (i.e. grammatical and ungrammatical), and acceptability measured varied pragmatic allowance (i.e. not acceptable at all/acceptable in local contexts). Both appropriateness and likelihood of feedback were then measured using a 5-point Likert scale, which is apt for opinion research. It is also suitable for this study because such scales are easy to construct, but yet yield reliable results (Kothari, 2008: 86). A sample of an item is shown in Figure 1.

Figure 1: Sample item from the questionnaire

Framework of analysis

Data from each category was processed in terms of frequency count, and was presented descriptively in percentages.

Findings

This study aims to explore how particular features of SgE are treated in the language classroom. In this section, overall results from the questionnaire are first described. They are then analysed and categorised to answer the research question in the next section.

Questionnaire results

Grammaticality and acceptability

Results for grammaticality and acceptability (Table 1) indicate that except for four features – lexical borrowings (e.g. Char Kway Teow), stative verbs to express actions (e.g. keep the book), unnecessary adverb/preposition (e.g. return back), and variant word connotations (e.g. spoilt electronics), most items were regarded as ungrammatical (range: 75%-95%), particularly items on grammar and syntax. Of the four, most participants were in high concordance that lexical borrowings and the use of stative verbs to express actions were grammatical. Interestingly, all items were generally marked as acceptable for use in the local context (range 60%-100%).

These results suggest that, firstly, the participants do not hold a prescriptive stance on SgE like traditionalists, but rather acknowledge its value and use in informal domains/local contexts; and secondly, participants have a good sense/knowledge^[7] of English grammar rules to distinguish these items as ungrammatical. This supplements findings of Schaetzel et al. (2010: 426) that trainee teachers are generally capable of distinguishing SCE items. However, whether they have knowledge of systematic differences between SCE and SSE is unascertained; finally, these features of SgE has stabilised extensively as norms, and are reaching considerable homogeneity, especially in their acceptability (Schneider, 2003: 266).

Feature	Grammaticality (%)			Acceptability (%)
	Yes	No	Unsure	Yes	No
Lexical borrowing	95	5	0	100	0
Stative verb to express dynamicity	85	10	5	95	0
Unnecessary adverb/preposition	60	40	0	95	5
Variant word connotation	55	45	0	95	5
No backshifting	25	75	0	85	0
Number disagreement	25	75	0	60	40
Pluralising non-count nouns	15	80	5	75	30
Discourse particle	10	90	0	95	5
Topic prominence	5	85	10	75	15
Object omission (topic chaining)	5	90	5	75	20
OSV structure	5	95	0	75	25
'Be' omission	5	95	0	70	30
Present tense to mark past tense	5	95	0	80	20

Table 1: Grammaticality and acceptability

Appropriateness for use by students in the classroom

Results for appropriateness (Table 2) show that perceptions of grammaticality affect appropriateness of use in the classroom, as items that were deemed ungrammatical in Table 1 were also regarded as inappropriate for the classroom. Likewise, most items, except for the same four features described above, were regarded as inappropriate (range: 5%-35%). This concurs with Saravanan and Poedjosoedarmo's (1997) findings that only SSE, or in this context grammatical English, is generally regarded as appropriate for the classroom by teachers. Responses with 'neutral' were also higher in frequency than that of 'unsure' in their returns of grammaticality and acceptability. This may imply that assessment of appropriateness may be more complicated than the notions of grammaticality and acceptability, which agrees with Tey (2000), where teachers revealed that non-standard English is appropriate at informal classroom situations, but not absolutely appropriate because they are also regarded as impediment to the acquisition of standard English.

Feature	Appropriateness (%)
	Appropriate	Inappropriate	Neutral
Lexical borrowing	80	0	20
Stative verb to express dynamicity	75	0	25
Variant word connotation	65	15	20
Unnecessary adverb/preposition	55	15	30
No backshifting	35	30	35
Discourse particle	35	10	55
Pluralising non-count nouns	25	45	30
Number agreement	20	50	30
Object omission (topic chaining)	20	50	30
Topic prominence	20	50	30
Present tense to mark past tense	15	50	35
OSV structure	10	40	50
'Be' omission	5	70	25

Table 2: Appropriateness for use by students in the classroom

Likelihood of corrections

Results of likelihood of corrections (Table 3) echo findings by Farrell and Tan (2007), where teachers do not believe that all instances of SgE regarded as ungrammatical should be corrected. In fact, generally, the likelihood of correction was unexpectedly low (with 9 features below 60%). Using the present tense to mark the past and 'be' omission were most likely to be corrected (70% for both), and with lexical borrowings and using a stative verb to express actions as most unlikely to be corrected (0% for both), which are in tandem with their high frequencies of grammaticality in Table 1. Again, it is interesting to note that there are high counts for 'neutral', which could be attributive of the dilemma described earlier. A caveat for these findings is that it does not imply definite translation to practice, and a measurement like such involves also beliefs. Nevertheless, it does bear indicative value.

Feature	Likelihood of correction (%)
	Yes	No	Neutral
Present tense to mark past tense	70	20	10
'Be' omission	70	0	30
Topic prominence	65	20	15
Pluralising non-count nouns	60	35	5
Number agreement	55	25	20
OSV structure	55	10	35
Object omission (topic chaining)	50	25	25
No backshifting	40	35	25
Discourse particle	35	40	25
Unnecessary adverb/preposition	20	60	20
Variant word connotation	15	65	20
Stative verb to express dynamicity	0	60	40
Lexical borrowing	0	70	30

Table 3: Likelihood of correction

Analysis and categorisation

In the analysis of trends and patterns, three groups were derived from the results gathered.

Group 1 – Grammatical and will not be corrected

Group 1 consists of two SgE features (Table 4): lexical borrowings that serve to identify local objects and cultural constructs which have no English equivalents, such as 'Char Kway Teow' (Hokkien) and 'Rojak' (Malay), and the use of stative verbs to refer to actions, such as keep to refer to 'put away' or wear to refer to 'put on'. These features have not only indigenised as an endornomative norm, but could have formalised to become acceptable in the formal domain of the classroom.

Feature	Grammaticality (%)		Acceptability (%)		Appropriateness (%)		Likelihood of correction (%)
	Yes	No	Yes	No	Yes	No	Yes	No
Lexical borrowing	87.5	5	92.5	0	77.5	0	0	70
Stative verb to express dynamicity	80	7.5	87.5	0	65	0	0	57.5

Table 4: Group 1 items

Group 2 – Ungrammatical, but more unlikely to be corrected

Two other SgE features (Table 5) – unnecessary adverb/preposition, such as 'return back', or 'discuss about', and variant word connotations, such as 'renovate', 'spoilt', 'bluff'^[8] – are both clearly indigenised forms as evinced by the high acceptability of use within the local context, although more participants perceived them as ungrammatical than those in Group 1. Yet, the equal division in the perspectives of grammaticality and low likelihood of correction of these items may imply that these items have stabilised as endonormative norms to a large extent, and could eventually become formalised endonormative norms.

Feature	Grammaticality (%)		Acceptability (%)		Appropriateness (%)		Likelihood of correction (%)
	Yes	No	Yes	No	Yes	No	Yes	No
Unnecessary adverb/preposition	60	40	95	5	65	15	20	60
Word connotation	55	45	95	5	55	15	15	65

Table 5: Group 2 items

Group 3 – Ungrammatical, and likely to be corrected

Finally, the remaining 9 items (Table 6) were considered by participants to be highly ungrammatical (range: 75%-95%), and have higher likelihood of corrections (range: 35%-70%) than those in Groups 1 and 2. However, as with other features, these features were also regarded as indigenised norms, given the significantly high acceptability ratings vis-à-vis low grammaticality scores. Incidentally, unlike Groups 1 and 2, many of these features concern deviations from fundamental grammar and syntactic rules (e.g. OSV structure, and number disagreement), and thus it may be more difficult for these items to stabilise to the extent of becoming formalised norms.

Feature	Grammaticality (%)		Acceptability (%)		Appropriateness (%)		Likelihood of correction (%)
	Yes	No	Yes	No	Yes	No	Yes	No
No backshifting	25	75	85	0	35	30	40	35
Discourse particle	25	75	60	40	35	10	35	40
Pluralising non-count nouns	15	80	75	30	25	45	60	35
Number disagreement	10	90	95	5	20	50	55	25
Object omission (topic chaining)	5	85	75	15	20	50	50	25
Topic prominence	5	90	75	20	20	50	65	20
Present tense to mark past tense	5	95	75	25	15	50	70	20
OSV structure	5	95	70	30	10	40	55	10
'Be' omission	5	95	80	20	5	70	70	0

Table 6: Group 3 items

The results are presented in an overview chart:

Figure 2: Treatment of SgE features

Discussion

Beliefs, knowledge, and practice

In terms of teachers' belief, results show that participants hold a non-prescriptive stance towards SCE in the classroom, supporting findings found in Saravanan and Poedjosoedarmo (1997) and Teh (2000). However, although it is propounded that the acceptance of the basilectal variety in the classroom may present several benefits on the acquisition of the standard variety (Rubdy, 2007: 320; Kwek, 2005: 81; Teh, 2000: 30; Rickford, 2006), the effectiveness of it as a linguistic resource is reliant on teachers' careful management (e.g. through contrastive analysis). In this regard, the results gathered for the likelihood of corrections reveal that some teachers can be over tolerant, resulting in the probability of a neglect of some features that could otherwise influence their acquisition of the standard. After all, the fact that students need to be taught the standard code in the classroom cannot be negated. This is because, unlike SCE, which is acquired through everyday use, SSE requires formal instruction like any other standard code (Goh and Silver, 2006), and students need both SCE and SSE to partake effectively in all domains of the community. Therefore, acceptance (beliefs) has to be coupled with conscious management (practice), by being cognizant of how to manipulate the vernacular to achieve the central purpose of teaching the standard. This presents as an implication for teachers' practice. In the same vein, this also bears implications on teachers' knowledge. First, it is necessary for teachers to have a technical/practical awareness of the possible effects of SCE on the acquisition of SSE if the vernacular is not treated carefully. This includes knowledge of systematic differences between SCE and SSE/StdE as surfaced from past studies (Dixon, et al., 2012: 223; Tan, 2005: 171; O'Neill et al., 2010: 56). In this, although results from this study show that participants are sensitive to ungrammatical forms, it is not known whether participants relied on intuition or on sound grammatical knowledge. This can be further researched. Secondly, teachers need skills on how SCE can be used as a classroom device to aid in the acquisition of the standard, and whether or not teachers are capable of doing so has not been ascertained in past research, and can be further studied.

How much localised norms to accept?

The gradual proliferation of SCE in the classroom (or even features of SSE that have deviated from StdE, but are still regarded as grammatical in the local context (e.g. Tan, 2011: 16-17), as also evinced by this study, has caused debate on the variety of English which should be taught in the classroom. If it is right that localised norms are displacing these norms of StdE, or at least, are causing linguistic influence to some extent, it is questionable if the language learnt is indeed one that is internationally intelligible (Goh, 1999; Lee, 1999; Lee, 2011). The question therefore is how much should an endonormative norm (whether SCE or SSE) be accepted in the classroom, even if it is regarded as standard or acceptable in the local context, and how much must traditional StdE teaching remain. And, in the most extreme stance, should the proposition that only StdE be taught in the classroom be warranted, since SgE is already so pervasive in use in society that no teaching is required. Or perhaps, a more practical issue is whether students are able to efficiently style shift/code switch between these codes. In this, Chng (2003: 54) rightly summarises that:

The gap between the official agenda to have every Singaporean speak internationally intelligible English and the opposing desires of the average Singaporean to maintain Singlish as a discourse option results in an ideological conflict over what makes Singapore work and what makes its inhabitants Singaporean.

The SCE-SSE debate

Finally, a more pertinent issue is the justifiability of government's persistent action in eradicating the vernacular from the classroom, and in fact, from society. The results from this study corroborate the literature that regardless of the government initiatives to eradicate the vernacular in the classroom, SCE is clearly bound to exist, especially since SgE is steadily stabilising as an endonormative norm and is common in the learners' milieu, even in that of teachers. Instead of proscribing the vernacular, several scholars (e.g. Tan and Tan, 2008) suggest inculcating students with a sense of linguistic mutability to manage the vernacular and appreciate variation. As linguist David Crystal described about the changes in language curriculum in the United Kingdom, Singapore could achieve linguistic 'freedom' through

a rapprochement between the study of the standard language, which is so important for promoting universal intelligibility, and the study of non-standard language, which is so important for promoting local identity.
(Crystal 2006: 142-43, cited in Tan and Tan, 2008)

Limitations and suggestion for further studies

There are several limitations to this study. Firstly, each feature of SgE is only represented by an example, and therefore not representative of the feature, nor of SgE as an entirety. However, the results are indicative of likely heterogeneity that exists between different SgE features with regard to its place within the process of endornomative stabilisation. Secondly, the sample size of 20 participants is small, and therefore the results are not generalisable. Also, these findings are representative of trainees who are linguistically trained. More research should also be carried out on other types of teacher, especially on practising teachers who hold degrees in other disciplines. In view of this, it is worth noting that the findings from this article are merely exploratory, and aim to provide preliminary insights into the issue. Finally, as defined earlier, endornormative stabilisation involves the simultaneous process of accepting local norms and distancing the colonial variety. Since this study involves investigating SgE, in particular SCE, therefore, a study that involves the investigation the treatment of StdE and SSE features could be done.

Conclusion

The results of this study have provided an insight into how teachers' cognition (beliefs, knowledge, and practice) affect the treatment of SgE, especially features of SCE, and has shown that different features of SCE are treated differently in the three aspects of teacher cognition. It depicts the language classroom as non-prescriptive, where the basilectal variety is not entirely proscribed, and even shows signs of boundary-crossing into SSE. Indeed, natural language changes are in progress, as also described in Tan (2011: 16). In a bigger picture, therefore, it illuminates the constant stabilisation of SgE as an endornomative norm in the classroom context – gradually accepting local norms, and, at the same time, distancing from StdE to gain linguistic independence; probably, soon enough, SgE could be seen in the last stage of the dynamic model: differentiation.

---

Acknowledgements

This research was completed as fulfillment of an independent research module at the National Institute of Education, Nanyang Technological University, Singapore. I would like to express my gratitude to my supervisor, Dr Tan Siew Imm, for her advice, experience, and patience. I am also indebted to my friend, Jason Chan, who provided valuable feedback.

To God be the glory.

List of figures

Figure 1: Sample item from the questionnaire

Figure 2: Treatment of SgE features

List of tables

Table 1: Grammaticality and acceptability

Table 2: Appropriateness for use by students in the classroom

Table 3: Likelihood of correction

Table 4: Group 1 items

Table 5: Group 2 items

Table 6: Group 3 items

Appendices

Appendix A

SgE features

Feature	Item
Pluralising non-count nouns	Andy and I are going to the furniture store to purchase some new furnitures.
Number disagreement	The biggest timewaster are meetings.
Object omission (topic chaining)	This is the sashimi I was talking about. People eat raw, you know?
Topic prominence	Bird flu, lucky the government was fast.
'Be' omission	My other brother in National Service.
No backshifting	George said that he cannot come to the party.
Present tense to mark Past tense	I can run very fast last time.
OSV structure	Sashimi I like.
Unnecessary adverb/preposition	Syu can never return back to her country.
Lexical borrowing	I would like to have a plate of Char Kway Teow.
Variant word connotation	My phone is spoilt, and I need to fix it.
Stative verb to express dynamicity	Keep the book in your bag.
Discourse particle	This is not your book, lah!

StdE features

Feature	Item
that-clause	That she should forget me so quickly was rather a shock.
because-clause	Because I worked fast, I finished early.
Collective nouns as plural in number	The Singapore soccer team are winning.
Past perfect tense	I sold the laptop that I had had since National Service.
Subjunctive 'be'	I wish it were Saturday.
Absolute constructions	Weather permitting, we shall have an open-air concert.
'-s' spelling ending of nouns not plural	Fish and chips is the best dish on the menu
Relative 'that' to refer to a person	There is the man that we met last night.
Subject pronoun 'I' after dummy subject	It was I who ate the cake.
Subject pronoun after 'as' comparison	Alan is as tall as I.
a/an according to first syllable sound	He is an NUS professor.
'Anyhow'	The suitcases were flung anyhow.
'How come'	How come you are late?

Appendix B: Instructions

In this survey, you will be asked to judge the grammaticality, appropriateness for classroom, and the likelihood of correction for a list of items. Please read the definition for each component below before starting the survey.

Grammaticality

Using either your intuition or knowledge of English grammar, decide whether the item is grammatical. If you think that it is ungrammatical, decide the extent to which the item can be used in your own everyday speech. Indicate this using the respective check box.

Appropriateness for the classroom

To what degree would you regard the use of this item in the classroom by your students to be appropriate?

Likelihood of correction

What is the likelihood of the item being corrected by you as a teacher if this item were to be used by your students during lesson time?

Note: This is not intended to be a test of your command of English, but a survey of your views in what is grammatical, and what is not. Therefore, it is important that you do not refer to an external source to aid you in the survey. Your honest responses are needed. Thank you.

Notes

[1] Jasper Sim is a final-year student of the BA (Education) programme at the National Institute of Education, Nanyang Technological University, Singapore. His interests include language education, Singapore English, and phonetics.

[2] This does not mean the SgE (SSE or SCE) is linguistically substandard, and this is done for convenience of reference.

[3] Standard English, following Low and Brown (2005: 11), refers to a variety of English whose grammar, vocabulary, and orthography are similar to that used in established varieties of English, such as the UK and the USA, and used in formal or official domains.

[4] Examples of features (both SSE and SCE) include aspects of phonology (e.g. Lim, 2004; on lexical stresses, Low, 2002; on dental fricatives, Moorthy and Deterding, 2002), in syntax (e.g. on noun phrases, Wee and Ansaldo, 2004; on its aspectual system, Bao, 2005; on its verbal clusters, Fong, 2004), and in lexis (e.g. on discourse particles, Lim, 2007; on discourse and lexis, Deterding, 2007).

[5] Both Chng, and Rubdy (2001) provided a comprehensive compendium of the SCE-SSE debate, and the general anti-SCE arguments to the debate were: (1) The use of SCE hinders the acquisition of SSE (Chng, 2003: 56; Goh, 1999; Lee, 2001); (2) SCE is not understood by others, and therefore the use is disadvantageous (SGEM, 2000); and (3) students may not be able to successfully shift between SSE and SCE (Goh, 2000).

[6] The academic structure of the institution requires those of the Secondary School track to take on two ASs, while those in the Primary School track take on one AS. This is equivalent to the 'major/minor' system in other universities.

[7] 'Sense' and knowledge can be distinguished as: the first relating to intuition of what constitutes grammatical, while the latter concerns the application of learnt knowledge of English grammar.

[8] Refer to Low and Brown (2005: 80-86) for more examples and the variations in connotation.

References

 

 

To cite this paper please use the following details: Sim, J.H. (2014), 'A Feature-based Study of English Trainee Teachers' Treatment of Singapore English', Reinvention: an International Journal of Undergraduate Research, Volume 7, Issue 1, http://www.warwick.ac.uk/reinventionjournal/archive/volume7issue1/hong Date accessed [insert date]. If you cite this article or use it in any teaching or other related activities please let us know by e-mailing us at .